Reproduce “Trajectory Optimization of Chance-Constrained Nonlinear Stochastic Systems for Motion Planning Under Uncertainty” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Yashwanth Kumar Nakka, Soon-Jo Chung
A framework combines Generalized Polynomial Chaos expansion, Distributional Robustness, and Sequential Convex Programming to compute trajectories for chance-constrained stochastic nonlinear systems. This approach provides high-probability safety guarantees for autonomous robots operating under uncertainty, validated through simulations and hardware experiments on Caltech's M-STAR robotic spacecraft simulator.
Key insights
- Generalized Polynomial Chaos effectively transforms stochastic differential equations into higher-dimensional deterministic ODEs, capturing complex uncertainty characteristics beyond mean and covariance.
- Distributional Robustness allows for the derivation of provably conservative and convex approximations of chance constraints, providing strong safety guarantees without assuming specific probability distributions.
- The combination of these techniques with Sequential Convex Programming results in a theoretically sound and computationally tractable framework for complex chance-constrained motion planning.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.